Improved pipeline for reducing erroneous identification by 16S rRNA sequences using the Illumina MiSeq platform

• Published in: The journal of microbiology Vol. 53; no. 1; pp. 60 - 69
• Main Authors: Jeon, Yoon-Seong, Park, Sang-Cheol, Lim, Jeongmin, Chun, Jongsik, Kim, Bong-Soo
• Format: Journal Article
• Language: English
• Published: Heidelberg Springer-Verlag 01.01.2015; The Microbiological Society of Korea; Springer Nature B.V; 한국미생물학회
• Subjects: Accuracy; Bioinformatics; Biomedical and Life Sciences; Cloning; Computational Biology; cost effectiveness; DNA libraries; DNA Primers; Gene Library; genes; high-throughput nucleotide sequencing; High-Throughput Nucleotide Sequencing - economics; High-Throughput Nucleotide Sequencing - methods; Life Sciences; Metagenome; Metagenomics; Microbial activity; microbial communities; Microbiology; nucleotide sequences; nucleotides; pipelines; ribosomal RNA; RNA, Ribosomal, 16S - genetics; Sequence Analysis, DNA - methods; species identification; Systems and Synthetic Microbiology and Bioinformatics; Taxonomy; 생물학; South Korea; MiSeq; identification; 16S rRNA gene
• ISSN: 1225-8873; 1976-3794; 1976-3794
• DOI: 10.1007/s12275-015-4601-y

The cost of DNA sequencing has decreased due to advancements in Next Generation Sequencing. The number of sequences obtained from the Illumina platform is large, use of this platform can reduce costs more than the 454 pyrosequencer. However, the Illumina platform has other challenges, including bioinformatics analysis of large numbers of sequences and the need to reduce erroneous nucleotides generated at the 3′-ends of the sequences. These erroneous sequences can lead to errors in analysis of microbial communities. Therefore, correction of these erroneous sequences is necessary for accurate taxonomic identification. Several studies that have used the Illumina platform to perform metagenomic analyses proposed curating pipelines to increase accuracy. In this study, we evaluated the likelihood of obtaining an erroneous microbial composition using the MiSeq 250 bp paired sequence platform and improved the pipeline to reduce erroneous identifications. We compared different sequencing conditions by varying the percentage of control phiX added, the concentration of the sequencing library, and the 16S rRNA gene target region using a mock community sample composed of known sequences. Our recommended method corrected erroneous nucleotides and improved identification accuracy. Overall, 99.5% of the total reads shared 95% similarity with the corresponding template sequences and 93.6% of the total reads shared over 97% similarity. This indicated that the MiSeq platform can be used to analyze microbial communities at the genus level with high accuracy. The improved analysis method recommended in this study can be applied to amplicon studies in various environments using high-throughput reads generated on the MiSeq platform.